| import requests |
| import json |
| import os |
|
|
| API_URL = "https://api.opentyphoon.ai/v1/ocr" |
|
|
| def ocr_upload(file_path, api_key, |
| model="typhoon-ocr", |
| task_type="default", |
| max_tokens=16384, |
| temperature=0.1, |
| top_p=0.6, |
| repetition_penalty=1.2, |
| pages=None): |
|
|
| if not os.path.exists(file_path): |
| raise FileNotFoundError(f"File not found: {file_path}") |
|
|
| with open(file_path, "rb") as f: |
| files = { |
| "file": (os.path.basename(file_path), f) |
| } |
|
|
| data = { |
| "model": model, |
| "task_type": task_type, |
| "max_tokens": str(max_tokens), |
| "temperature": str(temperature), |
| "top_p": str(top_p), |
| "repetition_penalty": str(repetition_penalty) |
| } |
|
|
| if pages: |
| data["pages"] = json.dumps(pages) |
|
|
| headers = { |
| "Authorization": f"Bearer {api_key}" |
| } |
|
|
| response = requests.post(API_URL, files=files, data=data, headers=headers) |
|
|
| if response.status_code != 200: |
| raise Exception(f"OCR API Error {response.status_code}: {response.text}") |
|
|
| return parse_ocr_response(response.json()) |
|
|
|
|
| def parse_ocr_response(result_json): |
| texts = [] |
|
|
| for page in result_json.get("results", []): |
| if page.get("success"): |
| content = page["message"]["choices"][0]["message"]["content"] |
|
|
| try: |
| parsed = json.loads(content) |
| text = parsed.get("natural_text", content) |
| except: |
| text = content |
|
|
| texts.append(text) |
| else: |
| print(f"❌ Error: {page.get('error')}") |
|
|
| return "\n".join(texts) |
|
|
|
|
| |
| |
| |
| if __name__ == "__main__": |
| api_key = "sk-SJLOZn2kGqDX2zuO9doBAwb1alfocOLYQSM9SkiU2HW3KpQC" |
|
|
| file_path = input("📂 ใส่ path ไฟล์: ").strip() |
|
|
| text = ocr_upload(file_path, api_key) |
|
|
| print("\n===== OCR RESULT =====\n") |
| print(text) |
